INDEX

Explanations

a specific repeated character or symbol

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

abi

-0.16

orde

-0.16

ÑĪÑĤ

-0.15

serter

-0.15

ypad

-0.14

Ð°ÑĢÐ°ÑĤ

-0.14

ivating

-0.14

oupper

-0.14

ghi

-0.14

Pub

-0.14

POSITIVE LOGITS

 Bomb

0.18

.sys

0.17

bomb

0.17

ãĥªãĤ¹

0.16

Bomb

0.16

 bomb

0.15

ipt

0.15

 Jones

0.15

 sleeper

0.15

 bombing

0.15

Activations Density 0.010%

a specific repeated character or symbol

No Comments

No Known Activations